PROJECT SUMMARY Radical cystectomy is the cornerstone of treatment for patients with muscle-invasive bladder cancer and is among the most complicated oncologic procedures performed in the United States. Along with neoadjuvant chemotherapy, these two modalities comprise the current standard of care for the disease. Approximately 1/3 of patients who undergo radical cystectomy after neoadjuvant chemotherapy will have no residual disease left in the bladder. Retrospective literature supports the use of cystectomy deferral in patients who undergo endoscopic biopsy showing no residual disease. However, this algorithm is prone to a high false negative rate necessitating salvage cystectomy, delay of which may result in worse survival outcomes. The first aim of the proposal will explore the phenomenon of mutation clearance and persistence in urine-derived cell-free DNA as a marker for disease absence or persistence, respectively, after endoscopic resection and chemotherapy. In preliminary studies, combination of single molecule tags and next generation sequencing resulted in high sensitivity detection of somatic variants previously identified in tumor whole exomes in prechemotherapy urine samples. At least one variant persisted in postchemotherapy urine samples from two of two nonresponders, and no variants were detected in urine samples from two of two complete responders. This preliminary data shows that the assay is feasible, reliable, and merits further exploration. The assay will be used on banked samples collected on a previously completed trial (NCT01031420) and on samples from a prospective clinical trial opening now at Fox Chase Cancer Center. Localized bladder cancer demonstrates a relatively high rate of pathological complete chemoresponse, but the mechanism of chemosensitivity remains incompletely understood. Data described herein suggests that the immune infiltrate is driven by CD8+ cells and high neoantigen density. Neoantigens are the non-self peptides that result from somatic missense or frameshift mutations that can be recognized by the immune system. Aim 2 of the proposal seeks to clinch a role for the immune system in this context by identifying tumor-specific cytotoxic T-cells which recognize tumor neoantigens. Expressed neoantigens will be identified using whole exome sequencing and RNA-seq of tumors and available computational algorithms. These mutant peptides will then be synthesized and used to screen live patient-derived CD8+ cells for neoantigen- specific cytotoxic responses. Samples to be used in this study will be obtained from a prospective clinical trial opening now at Fox Chase Cancer Center. This hypothesis-driven molecular and cellular biological study can potentially identify a new and unexpected mechanism for chemoresponse mediated by the immune system. If successful, these aims will go hand-in-hand in the future to prospectively and retrospectively identify responders and provide a framework to manipulate the immune system to achieve chemoresponse with the long term goal of safe radical cystectomy avoidance.